The focus of this research effort is on the generation of mathematical and statistical procedures applicable to the problem of low dose extrapolation in toxicological experiments involving endpoints other than carcinogenicity. Emphasis is given to the development of optimal procedures for ascertaining the presence of litter effects in laboratory data and to the extension of the Prentice generalized dose-response curves to toxicological experiments involving litter effects. The results of extrapolation based on biological theory are to be compared to those base on empirical models.